He was pointing at the screen, sweat beading faintly above his lip, which I only noticed because the meeting room was too cold-always 18.2 degrees Celsius, year-round, regardless of the season. A small, anxious detail that kept catching my attention while the big, deliberate lie unfolded.
“Engagement is up 8.2%,” the VP of Product declared, slamming his hand lightly on the table, generating a hollow, wooden thud that resonated with the hollowness of the statement itself. The arrow was aggressively green, pointing toward the upper-right corner of the chart, the universal sign that everything is fine, the signal that the emperor is fully clothed, and, most importantly, that we can all stop thinking now.
Someone, bless their heart, murmured the unavoidable question: “What exactly are we defining as ‘Engagement’ this quarter?”
The Truth of Inadmissibility
Silence. Not an intentional, dramatic silence, but the kind that happens when the entire apparatus of corporate data collection slams suddenly, violently, into the reality that nobody actually defined the word before they started optimizing for it. Engagement, it turned out, meant that 2,042 people had scrolled past the new feature tile without clicking it, but they had scrolled past it slower than they had before. Slow scrolling was the proxy for interest. The green arrow was nothing more than a measure of minor digital hesitation.
I’ve been the VP in that room. I’ve desperately pointed at the chart. I know the feeling. Data is supposed to be objective truth, but for most large organizations, it has become a security blanket-or worse, a pre-written apology. It’s proof that you were doing something, even if that something had no strategic foundation. We spend millions on BI tools that deliver thousands of metrics, and yet, the core frustration remains: we have a million dashboards, but nobody knows what the hell is actually happening. We’re suffering from the Delusion of Certainty.
It reminds me of the fight I had last week trying to return a $52 defective item without a receipt. The item was clearly broken. The staff agreed it was broken. But because I couldn’t produce the piece of heat-sensitive paper, the transaction-the truth-was inadmissible. The system required documentation, even if the truth was standing right there, visibly cracked, in front of the cashier. Corporate data systems are exactly the same way. The truth might be visible-low morale, failing product-market fit, a toxic culture-but if it isn’t captured on the approved 232-point dashboard, it doesn’t exist, and therefore, we can’t fix it.
OUTSOURCED CRITICAL THINKING
→ Algorithm Output ≠ Strategy
We have outsourced critical thinking to algorithms, and then we confuse the algorithm’s output with strategy.
The Food Stylist’s Lesson
This phenomenon is particularly apparent when you look at professionals who deal explicitly in presentation and sensory deceit. Take Hiroshi N., the food stylist I met several years ago in Tokyo. Hiroshi’s job is to make food look more delicious than it actually is for the camera. His entire expertise rests on understanding the gap between visual data and physical reality. His measurements are meticulous: the precise 0.02mm curl of a dried bonito flake, the $272 spent on lighting gels to make a piece of salmon look “freshly wet” when it’s actually been under studio lights for 4 hours. He is a master of creating an illusion that is measured and precise.
“A photo of food that is truly delicious rarely looks as good as a photo of food that is purely styled.”
Hiroshi N., Food Stylist
Think about that. We are modeling our entire corporate decision framework on Hiroshi N.’s methodology. We are styling the results for the stakeholders (the camera) rather than ensuring the underlying reality (the meal) provides genuine nourishment. We are optimizing for the metric that looks best under the $272 lighting, not the metric that signals sustainable, lasting growth. Our engagement rate isn’t measuring delight; it’s measuring visual optimization.
The Courage to Redefine
The contrarian angle here is simple, yet terrifyingly hard for organizations to swallow: Being ‘data-driven’ isn’t about having data. It’s about having the intellectual courage to look at the data and say, “This green arrow is meaningless, and we need to redefine what success looks like, even if it delays our next funding round.”
Most companies just use data to confirm existing biases. The CEO wants to launch Feature X. The dashboard shows that the small cohort of users who accidentally stumbled upon Feature X showed a slight behavioral change (scrolling slower, perhaps!). That data is immediately labeled ‘Proof,’ and the strategic decision is rubber-stamped. Data stops being a tool for inquiry and starts being a mechanism for accountability evasion. It allows leaders to say, “The data made us do it,” absolving them of the responsibility for strategy failure.
But real value is clear. When data connects directly to a strategic choice that benefits the future, the metrics clean themselves up. That clarity is why certain commitments, like choosing genuinely sustainable materials over single-use plastics, resonate so deeply. The impact is not a metric cooked up in a spreadsheet; it’s physical, measurable change. You don’t need a PhD to see the benefit. Companies like iBannboo understand that true metrics connect directly to value, not just clicks.
Precision vs. Accuracy
The mistake I made over the receipt, and the mistake companies make every day, is assuming that the paper trail is the same thing as the inherent value. They are not. If I spend $12,022 on a software integration that promises to increase my ‘customer satisfaction score’ by 2.2 points, I have to be willing to ask: What behavior change does that 2.2 points represent? Does it mean customers are genuinely happier, or did the survey just auto-fill the highest score if they didn’t click ‘next’ within 5.2 seconds?
2.2
This cycle feeds on itself. When the leaders ask for better data, the team responds by creating more dashboards-20 new charts per month, easily-because producing data is easier than producing insight. The focus shifts from the difficult work of strategy and competitive differentiation to the easy, safe work of optimizing proxies. We are optimizing the styling of the photo rather than focusing on the quality of the meal.
The Definition of Uselessness
Precision
Accuracy
My personal frustration… is that we confuse precision with accuracy. We can measure the click-through rate to the fourth decimal place. That’s precision. But if that click-through rate is irrelevant to the long-term health of the business, the metric is useless. It is precisely wrong.
We need to shift our focus from creating metrics that make us feel secure to creating metrics that force us to be honest. The most valuable dashboard might contain just two data points: the single metric that defines success and the single metric that defines failure. If the definition of ‘engagement’ forces a silent room for 32 seconds, then that definition is the strategic failure, not the trending line itself. We need to measure the hard truths, like the actual cost of employee churn, the cost of technical debt, or the time lost to pointless meetings-the things that erode the foundation, but rarely fit into the clean, green-arrow format.
The Final Question
Stop seeking proof for your pre-existing decisions. Start seeking genuine intelligence. The question isn’t whether your data is accurate. The question, the only one that matters, is this: What story is your data refusing to tell?
The Only Metric That Matters
What story is your data refusing to tell?
(If the answer is silence, you have found your next strategy.)